[pulseaudio-tickets] [Bug 103103] New: The client has not a chance to provide data, when absolute offset is used in a stream without prebuffering and underrun happend for long time

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=103103

            Bug ID: 103103
           Summary: The client has not a chance to provide data, when
                    absolute offset is used in a stream without
                    prebuffering and underrun happend for long time
           Product: PulseAudio
           Version: unspecified
          Hardware: Other
                OS: All
            Status: NEW
          Severity: normal
          Priority: medium
         Component: clients
          Assignee: pulseaudio-bugs at lists.freedesktop.org
          Reporter: mkid.dev at gmail.com
        QA Contact: pulseaudio-bugs at lists.freedesktop.org
                CC: lennart at poettering.net

I suspect that a problem, which I reported in GStreamer's pulsesink
(https://bugzilla.gnome.org/show_bug.cgi?id=788529) can have a source with
improper behavior of PulseAudio. Thus I would like to ask for clarification.

The problem can be simplified as follows:
a. a client stops providing samples (PulseAudio starts to render "silence",
underrun happens),
b. after some time (30 seconds) a client starting providing samples and write
them till is blocked (after samples for about 12 seconds are written) due to no
space left in client buffer,
c. after 18 seconds a client is woken and write samples into new offset, i.e 30
seconds from point b,

It seems that when absolute offset is used in a stream without prebuffering,
count of data requested during underrun has to be rendered, but a client
doesn't have a change to write all the data.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/pulseaudio-bugs/attachments/20171004/a9b85c43/attachment.html>